www.advmat.

de
www.MaterialsViews.com

REVIEW
Magnetic Materials and Devices for the 21st Century:
Stronger, Lighter, and More Energy Efficient
Oliver Gutfleisch,* Matthew A. Willard, Ekkes Brück, Christina H. Chen,
S. G. Sankar, and J. Ping Liu

1. Energy Consumption and the

A new energy paradigm, consisting of greater reliance on renewable energy Need for High-Efficiency Devices
sources and increased concern for energy efficiency in the total energy life-
Human development has caused a deple-
cycle, has accelerated research into energy-related technologies. Due to their tion of natural energy resources and
ubiquity, magnetic materials play an important role in improving the effi- climate changes with unpredictable conse-
ciency and performance of devices in electric power generation, conditioning, quences. New energy concepts are required
conversion, transportation, and other energy-use sectors of the economy. for the future of our industrial society,
resulting in an ever-increasing emphasis
This review focuses on the state-of-the-art hard and soft magnets and
on improving the efficiency of electricity
magnetocaloric materials, with an emphasis on their optimization for energy transmission and utilization and in the
applications. Specifically, the impact of hard magnets on electric motor and progressive replacement of oil-based fuels
transportation technologies, of soft magnetic materials on electricity gen- in transportation by electric motors. In this
eration and conversion technologies, and of magnetocaloric materials for context, functional magnetic materials,
refrigeration technologies, are discussed. The synthesis, characterization, and such as advanced hard and soft magnets,
magnetic refrigerants, magnetic MEMS
property evaluation of the materials, with an emphasis on structure–property
(microelectromechanical systems), mag-
relationships, are discussed in the context of their respective markets, as well netic shape memory alloys, and magneto-
as their potential impact on energy efficiency. Finally, considering future bot- rheological fluids and elastomers, can have
tlenecks in raw materials, options for the recycling of rare-earth intermetallics a substantial impact on improving these
for hard magnets will be discussed. technologies. This review will show that
magnetic materials are essential compo-
nents of energy applications (i.e., motors,
generators, transformers, actuators, etc.) and improvement in
magnetic materials will have significant impact in this area, on
Dr. O. Gutfleisch
Institute of Metallic Materials par with many “hot” energy materials efforts (e.g., hydrogen
IFW Dresden, 01069 Dresden, Germany storage, batteries, thermoelectrics, etc.). This section will high-
E-mail: o.gutfleisch@ifw-dresden.de light some technology sectors where magnetic materials have
Dr. M. A. Willard a clear impact, including transportation, electric power genera-
Naval Research Laboratory tion and conversion, and refrigeration.
Code 6355, 4555 Overlook Ave., S. W.
Washington, DC 20375 – 5343, USA
Electric motors have been ubiquitous in driving the techno-
Prof. E. Brück
logical society for the past couple of centuries. Motors ranging
Faculty of Applied Sciences from a few watts to several hundred kilowatts have been widely
Delft University of Technology employed in office and household appliances, the transporta-
2629 JB Delft, The Netherlands tion sector, and industrial drives. In developed countries such
Dr. C. H. Chen as the USA, nearly 30% of this electricity is consumed by the
Magnetics Laboratory industrial sector and, of that, nearly 65% is consumed by elec-
University of Dayton
Dayton, OH 45469, USA tric motor drives.[1] Thus the consumption of electricity energy
Dr. S. G. Sankar by motors, both big and small, in the USA alone, accounts for
Advanced Materials Corporation nearly 750 billion kWh or, at an average price of $0.06 per kWh,
850 Poplar Street, Pittsburgh, PA 15220, USA nearly 45 billion US dollars. A 1% improvement in efficiency
Prof. J. P. Liu would result in the savings of hundreds of millions of dollars
Deptartment of Physics and a reduction in CO2 emissions of nearly 2.2 million metric
University of Texas at Arlington
Arlington, TX 76019, USA
tons of carbon equivalent. Even a small improvement in energy
efficiency for electric motors can have a large economical and
DOI: 10.1002/adma.201002180 environmental savings. This is where permanent magnets and

Adv. Mater. 2011, 23, 821–842 © 2011 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim wileyonlinelibrary.com 821
 www.advmat.de
Table 1. World Total Net Electricity Consumers. Reproduced from.[2]
REVIEW
Country Electricity Consumed (Trillion kWh)
USA 3.82
European Union 2.93
China 2.53
Japan 0.98
Russia 0.82
Canada 0.53
India 0.52
Other Countries 4.25
Total World Consumption 16.38
soft magnets play a major role by providing a smaller, lighter
weight electric motor with the necessary power and torque
required for transportation and actuation applications.
Electricity consumed by the entire world population is esti-
mated to be about 16.38 trillion kWh (2006 estimates) broken
down by region, as shown in Table 1.[2] People in the USA con-
sumed 4.2 trillion kWh of electricity in 2007, a number that is
expected to grow from today’s 33% of our total energy consump-
tion to 50% by 2015.[3] Modern lighting, consumer electronics,
and transportation consistently require higher levels of power
quality and more total electrical power. This is irrespective of
an increase in the green energy sectors of electric power gen-
eration (see Figure 1 (left)). The need for efficient generation,
transmission, and distribution of electric power is ever growing,
however, the trend in efficiency has shown an alarming devel-
opment (Figure 1 (right)), with annual electric losses outpacing
annual increases in electricity consumption by a factor of more
than two-to-one (49.4 and 23.4 TW per yr, respectively). As a
result, while more energy can be produced from renewable
sources, electrical losses must also be addressed.
With the anticipated increase in demand for electric trans-
portation (including hybrid and electric vehicles, HEVs and
EVs), the consumption of electricity will increase significantly
above its current 2.5% annual increase. Therefore, it is neces-
sary to offset at least a part of the increased consumption of
electricity through innovative design and development of high-
efficiency motors, generators, and power converters. However,
improving the efficiency of the existing infrastructure will not
be enough to address the increased demand for electricity,
which has driven the acceleration of research and development
Figure 1. (left) Electric power generation in the United States by source for 2007. (right) Residential electricity consumption since 1949. Electric losses
include energy lost from generation, transmission, and distribution. Data from Ref. [4].
822 wileyonlinelibrary.com © 2011 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
www.MaterialsViews.com
Oliver Gutfleisch graduated
in 1991 in Materials Science
at the Technical University of
Berlin, Germany. In 1995 he
earned his Ph.D. at the School
of Metallurgy and Materials,
University of Birmingham,
UK, and continued his work
there on rare-earth transition
metal compounds for perma-
nent magnet applications as
a research fellow. He joined
the Leibniz Institute of Solid State and Materials Research
(IFW Dresden, Germany) in 1998, leading a research
group on functional materials for sustainable energy. In
2007 he earned his habilitation at the Technical University
of Dresden. He currently is a visiting professor at the
Imperial College London and a visiting advisor at NIMS
Tsukuba, Japan. In 2011 he is a Distinguished Lecturer
of the IEEE Magnetics Society for “Magnetic Materials in
Sustainable Energy”.
of clean and renewable energy sources. Renewable or sustain-
able energy refers to the conversion of solar, wind, ocean, hydro-
power, biomass, and geothermal energy into electric power.
The largest electric consumptions for the residential market
in the USA are refrigeration for food and air conditioning, which
contribute over 25% of the total for the average household.
Conventional cooling methodologies using vapor-compression
refrigerators prove to be bulky, heavy, and energy inefficient.
Recently, the use of the adiabatic demagnetization (the so-called
magnetocaloric effect (MCE)) for room-temperature refrigera-
tion was demonstrated to have ≈30% better energy efficiency
than competing vapor-compression refrigeration techniques,
without the environmentally detrimental refrigerant gases.[5]
Advances in MCE materials will enable solid-state technologies
to improve refrigeration capacities and efficiencies.
Whether referred to as sustainable, renewable, or alterna-
tive energy, a large number of new sources for the generation
of electricity are likely to be added to the grid in the next
decade.[6,7] This will require a strong investment in energy effi-
cient methods for power generation. In some cases, such as
wind power and transportation, improvements demand not
only higher efficiency but also smaller size and weight of the
Adv. Mater. 2011, 23, 821–842

www.MaterialsViews.com
materials and devices. Equally important to sustainable invest-
ment in electricity sustainability is developing higher energy
efficiency for power conversion. Advanced permanent and soft
magnet materials provide high efficiency and reliability and
compact, low cost, and low maintenance solutions for renew-
able energy technologies, including wind turbines, hydroelec-
tric power generators, and wave power buoys.
2. Hard Magnetic Materials
During the twentieth century, several permanent magnet mate-
rials were discovered. Techniques to effectively manufacture
these magnets have been established.[8] Device designs using
such magnets in various active and passive applications have
been successfully exploited. The energy product, which is a
key figure of merit of permanent magnets, has been enhanced,
starting from ≈1 MGOe for steels discovered during the early
part of the century, increasing to ≈3 MGOe for hexagonal fer-
rites, and finally peaking at ≈56 MGOe for neodymium-iron-
boron magnets during the past ten years. With this, almost
90% of the limit for the energy density, (BH)max, (based on the
Nd2Fe14B phase) can be achieved in commercially produced sin-
tered Nd-Fe-B grades. The historical evolution, spanning nearly
100 years, of these permanent magnets is shown in Figure 2.
At the time of writing, however, it appears that the search for
novel hard magnetic compounds with higher remanent mag-
netization, defined now as the relevant parameter, has some-
what stagnated and no further breakthrough is in sight. On the
other hand, only a small number of ternary and quaternary sys-
tems have been investigated so far. The approach of nanocom-
posites is the one that is currently the most actively pursued,
with particular emphasis on attempting to have the hard mag-
netic phase textured, as well as exchange-coupled with a soft
magnetic phase; the latter phase has an intrinsic upper limit of
Figure 2. Development in the energy density (BH)max at room tempera-
ture of hard magnetic materials in the 20th century and presentation of
different types of materials with comparable energy density (each magnet
is designed so that at a reference point 5 mm from the surface of the pole,
a field of 100 mT is produced).
Adv. Mater. 2011, 23, 821–842 © 2011 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
www.advmat.de
μ0Ms = 2.43 T for an Fe65Co35 alloy, where μ0 is the permeability
REVIEW
of free space and Ms is the saturation magnetization.
Recently, there is a much revived interest in various types
of high-performance permanent magnets based on rare-earth
intermetallic compounds. This is triggered by, for example, the
growing demand for energy efficient technologies in which
these magnets often play a pivotal part. Generally, the major
driving force for research and development of rare-earth perma-
nent magnets (RPMs) is the need for maximized energy densi-
ties at various operating temperatures. Most importantly, this
includes less Dy-containing Nd2Fe14B-type magnets with much
improved temperature stability for electromotor applications at
around 450 K,[9] Pr2Fe14B-type magnets for applications at 77 K
together with high Curie temperature (Tc) superconductors,[10]
and a new generation of SmCo 2:17-type magnets for applica-
tions at temperatures exceeding 670 K.[11,12] It also includes
magnetic-power MEMS,[13–16] for example, a high-speed perma-
nent magnetic generator, which require highly textured, thick
RPM films.[17] Nowadays, emphasis of research is on the con-
trol of microchemistry and structure of grain boundary phases
and internal interfaces, which are crucial for the understanding
of the relevant coercivity mechanisms and the related critical
elementary magnetization reversal processes. Textured nano-
composites could be the next generation of permanent mag-
nets. This would involve mastering the fabrication of intimately
mixed multiphase and well-oriented nanoscale magnets, which
cannot be done by conventional processing techniques.
2.1. General Considerations for Permanent Magnetic Materials
in Energy Applications
Permanent magnetic materials used in electric machines per-
form one basic function, which is to provide magnetic flux. This
function requires a saturation magnetization (4πMs in cgs units
or Js = μ0Ms in SI units) as high as possible and affordable, as
well as an appropriate coercivity. Figure 3 shows the saturation
magnetization versus intrinsic coercivity, iHc, for major soft and
hard magnetic materials. Most of the soft magnetic materials
have iHc << 1 kA m−1 and the hard magnetic materials have iHc
up to ≈2800 kA m−1 (≈35 kOe).
For permanent magnets, coercivity is the ability to resist
demagnetization, including field demagnetization from the
electric or magnetic circuit and thermal demagnetization from
the operating temperature. Coercivity is extremely important
for motor applications since these two demagnetization actions
occur continuously while motors are running. Two coercivity
parameters are used to grade the magnetic hardness: one is
intrinsic coercivity, iHc (or jHc), and the other is normal (or
technical) coercivity, Hc. The permanent magnets suitable for
such applications must have high coercivity and ideally iHc is
much higher than Hc, which allows the magnets to have linear
demagnetization induction (B) curve in the second quadrant.
Linear B curve is a very important characteristic that enables
the magnets to be stable during operation. Only three types of
permanent magnets shown in Figure 3 possess such capability:
these are hard ferrites, Nd-Fe-B, and Sm-Co magnets. Four pairs
of demagnetization curves at room temperature are shown in
Figure 4 for those three magnets and for an Alnico magnet. As
wileyonlinelibrary.com 823
 www.advmat.de
www.MaterialsViews.com
REVIEW

Figure 3. Polarization versus coercivity of soft and hard magnetic materials. The inset shows a closer look at high energy density NdFeB-type magnets
with various Dy-contents for applications from room temperature to 200 °C.

shown in Figure 4, Alnico has nonlinear demagnetization curves
and it is not suitable for motor applications. Sm-Co type mag-
nets have a very good, linear B curve at room temperature and
retain the linear B curve in the temperature range of 300°C to
550 °C, depending on the alloy composition.[11,12,18,19] Nd-Fe-B,
plotted in Figure 4(left), is not a good candidate for motor appli-
cations since a nonlinear B–H curve with a “knee” is shown on
its B curve, which indicates instability of this magnet when it is
surrounded in the electric field at elevated temperature. High
coercivity (Nd10Dy4)Fe80B6 magnets with 4 at% Dy substitu-
tion for Nd were developed for such applications. As shown in
Figure 4(right), the Dy added Nd-Fe-B retains linear B curves at
temperatures up to about 175 °C. Most of commercially avail-
able hard ferrites do not have a linear B curve at room tempera-
ture; however, that does not prevent ferrite from being used for
motor applications since ferrite has a unique positive tempera-
ture coefficient of coercivity, as shown in Figure 5.[20] The iHc of
the ferrite magnet increases from 3.2 kOe at 0 °C to 5.9 kOe at

Figure 4. Demagnetization curves in the 2nd quadrant for four major permanent magnets (left) and at various temperatures for Nd-Dy-Fe-B used for
motors (right). The (BH)max at 25 °C is in the parenthesis. Reproduced with permission.[21]

824 wileyonlinelibrary.com © 2011 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim Adv. Mater. 2011, 23, 821–842
 www.advmat.de
www.MaterialsViews.com

REVIEW
Figure 5. Magnetic properties of four magnets at up to 300 °C: energy product (BH)max versus T (left) and coercivity iHc versus T (right).[20] The (BH)max
at 25 °C is in the parenthesis. Reproduced with permission.[20] Copyright 2000, IEEE.

300 °C, making ferrite a relative good material for motor appli-
cations. With the exception of ferrite, all other magnets shown
in Figure 5 have negative temperature coefficients of coercivity.
There are other considerations for permanent magnetic
materials for energy related applications, including the load
line for the magnets. The load line, also called the permeance
coefficient, is only related to the geometry of the magnet. Figure 6
shows the demagnetization curves at 25 to 200 °C in 25 °C
increments for an Nd-Dy-Fe-B magnet that can be used for
hybrid vehicle motors. The figure also shows the load lines with
corresponding dimensions for cylindrical magnets. The load
line intersects with the B curve (B vs H), and the crossing point
of the two lines is the operating point of the magnet, Bd/Hd. Hd
is the demagnetization field and Bd is the induction B at Hd. The
operating points for different temperatures are listed in Table 2,
and the points in the shaded cells are not suitable for motor
applications since they are near the knees of the B–H curves,
where the magnet is vulnerable to demagnetization during
the dynamic operation. One example can be used to explain a
load line shown in Figure 6: the disc magnet with the length

Figure 6. Demagnetization curves at 25 to 200 °C for an Nd-Dy-Fe-B magnet and the load lines with corresponding dimensions for cylindrical magnets.
Reproduced with permission.[21]

Adv. Mater. 2011, 23, 821–842 © 2011 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim wileyonlinelibrary.com 825
 www.advmat.de
www.MaterialsViews.com

Table 2. The operating point Bd/|Hd| for each operating temperature.

REVIEW

Load line Operating point Bd/|Hd| (in kG/kOe)

|Bd/Hd| L/D 25 °C 50 °C 75 °C 100 °C 125 °C 150 °C 175 °C 200 °C
0.2 ≈0.15 1.80/9.4 1.76/9.1 1.72/8.8 1.7/8.7 1.6/8.2 1.5/7.9 1.3/6.9 1.1/5.8
0.5 ≈0.25 3.75/7.5 3.7/7.2 3.6/7.1 3.4/6.8 3.2/6.6 3.1/6.4 3.0/6.2 2.8/5.6
1.0 ≈0.4 5.7/5.7 5.6/5.6 5.4/5.4 5.2/5.2 5.1/5.1 5.0/4.9 4.8/4.8 4.7/4.7
2.0 ≈0.7 7.5/3.8 7.4/3.7 7.2/3.6 7.1/3.5 6.9/3.4 6.7/3.3 6.4/3.2 6.2/3.1

∗ The operating points in the shaded cells are not suitable for motor applications.

over diameter ratio (L/D) of 0.15 has a load line with |Bd/Hd| =
0.2, and the operating point is 1.5 kG/7.9 kOe at 150 °C. This
magnet is not stable when the operating temperature is higher
than 150 °C because the operating points are near or below the
knees of the B curves at 150 °C to 200 °C. For design engineers,
this thin magnet should not be used at temperatures higher
than 125 °C if safety is a factor under consideration.
2.2. World Market for Permanent Magnetic Materials
NdFeB magnets, in particular, have significantly impacted the
development of computer peripherals such as voice coil motors
and actuators, both in down-sizing and in enhancing their per-
formance characteristics. For example, in 1984, computer disk
drives of ≈10 MB size with random access time of over 60 ms
were the norm. Today, external hard drives of much smaller
physical dimensions but with a storage capacity of a few hun-
dred GB, running at speeds up to 7200 rotations per minute
(rpm) and with seek times of ≈8 to 12 ms have become common.
More importantly, these high-performance drives are available
at affordable costs. These extraordinary developments are partly
attributable to the availability of magnets with energy products in
excess of 50 MGOe. The production of NdFeB sintered magnets
has steadily increased from about 6000 tons in 1996 to about
63 000 tons in 2008 (see Figure 7 and Table 3).[22] The volume of
these magnets manufactured in China, Japan, USA, and Europe
together with their value (US dollars (US$)) and price during
the past decade are shown in Table 3. It is interesting to note
that almost 80% of this strategic material is actually produced in
China and manufacture in the US has ceased since 2004.
The world market for permanent magnets was estimated at
≈$8 billion in 2007 and is predicted to be $11 billion by 2010,
with 34% hard ferrite, 65% rare earth magnets, and 1% Alnico.
The growth rate of total permanent magnets is estimated to be
≈12% annually.[22] The fastest growing market is for Nd-Fe-B-
based magnets with high energy product, which would capture
62% market share in 2010. The dominant growing area of per-
manent magnetic materials is the energy-related applications.
Design, development, and manufacture of consumer elec-
tronics goods, such as DVDs, iPods, cameras, sensors, and cell
phones, are taking place at a rapid pace. All these devices include
components that utilize permanent magnets, such as the vibra-
tory motor in a cell phone. High-energy magnets enable the
design of these miniaturized devices. Such a miniaturization
strongly impacts the reduction in electrical power consumed in
the operation these devices, either from the mains or through
the rechargeable batteries. Thus, these devices now operate at
high energy efficiencies. Large quantities of NdFeB magnets
are used in the construction of battery-operated tools, magnetic
resonance imaging (MRI) units, electric bicycles (EB), electric-
assisted vehicles (EAV), speakers, magnetic separation units,

Figure 7. Output of NdFeB permanent magnets during the past ten years (left)[22] and predicted percentage sales ($) for 2010 of the major permanent
magnets in the world (right).[23]

826 wileyonlinelibrary.com © 2011 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim Adv. Mater. 2011, 23, 821–842
 www.advmat.de
www.MaterialsViews.com

Table 3. Output (in tons), sales value (in US$ × 108), and price (US$ per kg) of NdFeB sintered magnets in China, Japan, USA, and Europe from 2003

REVIEW
to 2008.[22]

2003 2004 2005 2006 2007 2008

China Output 15 550 22 910 30 160 38 220 45 100 49 880
Value 5.598 7.7894 9.9528 13.377 18.04 20.9496
Price 36 34 33 35 40 42
Japan Output 6200 7000 8500 10 500 11 800 12 600
Value 5.084 5.74 7.14 9.345 11.21 12.348
Price 82 82 84 89 95 98
USA Output 100 – – – – –
Value 0.08
Price 80
Europe Output 810 800 950 1080 1210 1100
Value 0.6561 0.648 0.779 0.9398 1.1253 1.067
Price 81 81 82 87 93 97
Global Output 22 660 30 710 39 160 49 800 58 110 63 580
Value 11.4181 14.1774 17.8718 23.1396 30.3753 34.3646
Price 50.4 46.2 45.1 46.47 52.3 54.1
China/Global 68.6% 74.6% 76.1% 76.7% 77.6% 78.5%
Japan/Global 27.4% 22.8% 21.5% 21.1% 20.3% 19.8%
USA/Global 0.4% – – – – –
Europe/Global 3.6% 2.6% 2.4% 2.2% 2.1% 1.7%

etc. This market segment is shown in Table 4 for China. The
market can be divided into three groups; the first one is termed
the high-tech; the second group as the traditional and the third
one as the low grade. It is interesting to note that the first group
has experienced a four-fold expansion while the second group
(traditional) has nearly doubled during the past five years.
Today ferrite magnets have the largest share in terms of
tonnage in energy-related applications, which is attributed to
their low price. Based on the sales data in Figure 7, we esti-
mate annual usage is ≈433 kilotons for ferrite magnets and
≈68 kilotons for Nd-Fe-B magnets. As described earlier, the
invention of ferrite in 1946 revolutionized the design of motors
and generators. Despite the low-energy product (less than 10%
of that of Nd-Fe-B magnets) today ferrite is still the choice
of many engineers because of its low price. For the motors
used for hybrid vehicles, Dy partially substitutes Nd in order

Table 4. Distribution of NdFeB magnet applications in China (in tons).[22]

Application 2003 2004 2005 2006 2007

MRI 360 560 760 1110 1800
VCM 200 350 530 995 1300
CD pick-up 840 1150 1540 2305 2515
DVD/CD ROM 1250 1800 2440 3290 4060
Mobile phone 360 680 950 1910 3160
Coreless tool 960 1250 1490 1990 3160
EB+EAV 1330 2280 3420 4560 5860
Subtotal (high-tech) 5300 8070 11 130 16 160 21 855
Speaker 3850 5730 7540 9560 11 280
Separator 2270 3420 4090 3950 3610
Magnetizer 1030 1070 1210 1150 990
Subtotal (traditional) 7150 10 220 12 840 14 660 15 880
Low-grade use 2550 3840 4980 6080 6320
Other 550 780 1210 1320 1045
Total 15 550 22 910 30 160 38 220 45 100

Adv. Mater. 2011, 23, 821–842 © 2011 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim wileyonlinelibrary.com 827
 www.advmat.de
REVIEW
to increase the coercivity (Hc) to a level sufficient for high-
temperature applications. The disadvantages of this are the
reduction in the energy density, (BH)max, caused by the antipar-
allel coupling between Dy and Fe in the (Nd,Dy)2Fe14B unit cell[24]
and the rarity of Dy.[9] It is estimated that currently 97% of Dy
rare-earth metal supply worldwide (1200 tons in 2008) is of
Chinese origin.[25] Metal prices in 2009 show that Dy costs ≈8
times as much as Nd metal.[26] In order to keep the cost as low
as possible and energy product as high as feasible, different
applications use different grades of Nd-F-B-based magnets, as
indicated in the insert of Figure 3. Common perception usually
is that Sm-Co is much more expensive than the Nd-Fe-B mag-
nets, since the Sm metal is more expensive than Nd and the
production process for Sm-Co is more complex than that for
Nd-Fe-B. The raw material cost of Nd-Dy-Fe-B is ≈$20 per kg,
higher than that for Sm-Co 2:17 metals, ≈$18 per kg, as listed
in Table 5. The longer manufacturing process for Sm-Co makes
its price ≈15% higher than Nd-Dy-Fe-B’s; however, if a higher
acceleration and a higher fuel efficiency (requires higher oper-
ating temperature) for certain electric cars are the goals, Sm-
Co could actually be an alternative (compare Figure 5 (left)).
Heavy rare-earth elements like dysprosium and to some extent
terbium are key components in producing RPMs that show the
required temperature stability for motor applications. Looking
at the supply situation, a strategy and novel microstructures
for RPMs are required to reduce the use of Dy. Another impor-
tant issue for the future is the recycling of RPMs, again to
ensure a stable supply of rare-earth metal sources. For this
purpose, hydrogen-assisted processing of RPMs could be very
advantageous.[27–29] The hydrogen decrepitation process (HD)[30]
could be used in the recycling of SmCo5, Sm2(Co,FeCu,Zr)17,
and NdFeB-type magnets. The growth of electric vehicles and
with increasing environmental legislation, which is likely to
demand better energy efficiency, and more resourceful use of
materials including recycling, especially for automotive applica-
tions where “end of life” rules are already being implemented.
Such measures can be justified in terms of environmental
impact (metal contamination of landfill sites), energy saving
(CO2 reductions), and resource depletion.
2.3. Torque and Efficiency
Gieras and Wing[31] point out that “the use of permanent mag-
nets in the construction of electrical machines result in the fol-
lowing benefits: (a) no electrical energy is absorbed by the field
excitation system and thus there are no excitation losses which
means substantial increase in efficiency; (b) higher torque
Table 5. Cost of raw materials of the two types of magnets (price based on July 2009 data).[26]
Nd-Dy-Fe-B
Nd Dy Fe
at% 10.00 4.00 80.00
wt% 21.77 9.81 67.44
Price ($) 4.4 14.2
Total cost ≈$20/kg
828 wileyonlinelibrary.com © 2011 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
www.MaterialsViews.com
and/or output power per volume (of the machine) than when
using electromagnetic excitation, and (c) better dynamic per-
formance than motors with electromagnetic excitation (higher
magnetic flux density in the air gap)”.
In addition to the developments in the permanent magnets,
there has been a substantial progress in semiconductor elec-
tronics resulting in the availability of switching devices, such as
MOSFETs (metal oxide semiconductor field-effect transistors)
and IGBTs (insulated-gate bipolar transistors), which are used
to commutate the motors. Further, developments in power elec-
tronics enable the design and manufacture of intelligent con-
trollers to operate electromechanical devices. Computer soft-
ware tools employing finite-element analysis techniques have
assisted in rapidly prototyping electromechanical devices. Thus,
the synergistic advances in these fields have enabled the rapid
growth of the permanent magnet motor industry.
The electromagnetic interaction between the magnetic flux
produced by a permanent magnet with a flux density, B, and
the electrical current flowing through a surrounding coil of
length, L, and area of cross section, A, carrying current with a
current density, j, results in the generation of torque τ = BjrAL,
where r is the length of the torque arm. Generally, the magnetic
flux is focused to interact effectively with the current through
the use of a soft magnet, such as soft steel, placed within the
available space envelope.[32] During operation of a motor, eddy
currents, hysteretic effects, and mechanical friction all con-
tribute to electric losses in the form of heat. The ratio between
the mechanical power, Pmech, derived from running the motor
at an angular velocity, ω, and the electrical power, Pelec, pro-
vided into the motor is the efficiency, η, of the motor. Thus, η =
(Pmech)/(Pelec) = (τω)/(VI). In this expression, V is the voltage
supplied to the motor that produces a current I. Torque versus
speed or torque versus current and the efficiency of the motor
describe the basic performance characteristics of a motor. The
design engineers have to a) maximize the flux density, B, by a
suitable selection of the permanent magnet as well as the soft
magnet material; b) maximize the interaction between the flux
lines of the magnet and the current flowing through the sur-
rounding wire by optimizing the geometry of the device; and
c) minimize the losses due to eddy currents, hysteresis, and
frictional effects and, if necessary, find an effective means to
withdraw the heat produced in the motor by e.g., mounting it
on a heat sink.
Gieras and Wing[31] compared the performance of a small
three-phase, four-pole, 1.5 kW, 50 Hz cage-induction motor
with that of a rare-earth permanent magnet (PM) brushless
motor. The full-load efficiency of the former is about 75% while
that of the latter is nearly 88%. Thus, the induction motor
Sm-Co (2:17)
B Sm Co Fe Cu Zr
6.00 13.3 59.6 19.3 5.7 2.1
0.98 28 49.1 15.1 5.1 2.7
5.9 9.0 0.5 0.5
≈$18/kg
Adv. Mater. 2011, 23, 821–842

www.MaterialsViews.com
Figure 8. Some advantages of a PM motor over an induction motor (IM). Reproduced with permission.[33]
draws nearly 2000 W of electricity and the PM motor draws
only 1704 W. The power savings is 296 W (or ≈15%) per motor.
Figure 8 shows some of the advantages of PM motor versus
induction motors.
Honkura[34] demonstrates that the enhancement in the
energy efficiency of a motor using rare-earth bonded magnet is
nearly 13% compared to a ferrite magnet motor, both operating
at the maximum 400 W. The power density of the rare-earth
permanent magnet motor was improved nearly twofold com-
pared to the ferrite motor.
Electric vehicles, hybrid as well as plug-in, employ brush-
less motors. Proper design of such motors and the control
electronics is now a very important area of worldwide research
and development. Key objectives are to a) design light-weight,
high torque-to-weight ratio motors, b) achieve high efficiency
throughout the operational range, c) provide reliable control,
d) have rugged construction, e) provide low maintenance
requirements, and f) supply the customers at low cost. It is also
important to note that electric motors have high torque at low
speeds. In contrast, in internal combustion engines the torque
increases with speed until reaching an optimum power level and
then falls sharply. Thus, the permanent magnet motors offer a
number of benefits in furthering this mobility application.
The possibilities of producing exchange-coupled perma-
nent magnets that offer the ability to exhibit energy products
in excess of ≈60 MGOe are being investigated by a number of
researchers around the world. However, in the case of conven-
tional brushless, slotted motors (which are typically employed in
Figure 9. World total installed wind energy during the last decade (left) and country share of total capacity 2009. Reproduced with permission.[36]
Copyright 2010, World Wind Energy Association.
Adv. Mater. 2011, 23, 821–842 © 2011 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
www.advmat.de
REVIEW
many applications), the torque levels off at about 40 MGOe due
to the flux saturating the soft magnet that is used in the core.
Simizu et al.[35] modeled the possibilities of designing brushless
motors that are slotless and demonstrated the overwhelming
advantages of such designs in the further enhancement of
torque using permanent magnets with potentially much
improved high-energy products. A nearly 25% increase in the
torque of slotless, brushless permanent magnet motors over
the slotted designs at about 60 MGOe, with a further poten-
tial for such increases at higher-energy products of permanent
magnets, have been predicted. Such designs will also result in
an increase in energy efficiency. Further advantages of adopting
high energy density magnets include higher torque-to-weight
ratio and reduction in noise and clogging.
2.4. Renewable Energy Generators: Wind Power and Magnetic
Materials
Wind power is the fastest growing sector within the renewable
power generation. According to the World Wind Energy Associa-
tion (WWEA), worldwide wind power capacity was 159 213 MW
in 2009. As shown in Figure 9, wind power capacity has doubled
every three years in the last decade and is predicted to be around
1 900 000 MW in 2020.[36] A typical commercial wind turbine
consists of three blades connected to an asynchronous gener-
ator via a gearbox, thus transforming the mechanical power into
electrical power. Asynchronous generators are constructed with
wileyonlinelibrary.com 829
 www.advmat.de
REVIEW
soft magnetic lamination and iron and copper armature wind-
ings with the excitation field provided by copper coils.
Various wind turbine concepts have been developed and built
to maximize the energy harnessed, to minimize the cost, and to
improve the power quality. Turbine concepts can be classified
with respect to the rotational speed, the power regulation, and
the generator system. When considering the construction of the
generator system, the turbines can be divided into direct-drive
and geared concepts. Generator systems can be separated into
the electrically excited (EE) machine and the permanent magnet
(PM) machine. The direct-drive generator system, especially
direct-drive PM generator system, is superior in terms of the
energy yield, reliability, and maintenance.[37] Because of reliability
issues and high servicing schedules of gearboxes, turbine pro-
ducers are developing direct-drive (DD) generators, which allow
power transformation without a gear. This has resulted in a shift
to synchronous generators because of the higher torque density
with the excitation field provided by the permanent magnet. New
designs use Nd-Fe-B permanent magnets in the synchronous
generators (PMSG). As shown in Figure 10, four permanent
magnet Nd-Fe-B generators are major components in advanced
wind turbines. A comparison of the new design with PM gen-
erators and a conventional generator yields drastic differences
in volume and weight[38] as well as higher operating efficiency,
higher torque density, 50% lower internal heat generation, and
easier assembly and maintenance.[39,40] Compared to a geared
system, the torque can be 100 times larger for a DD generator.
Permanent magnet generators should have more advan-
tages compared to conventional generators if the design of the
generators can be revolutionized by fully utilizing the high-
energy products of RPMs. The designs for the PM generators
today are more or less similar to the configurations devel-
oped for using hard ferrite magnets invented in 1946, which
revolutionized the design of inductors, transformers, motors,
and generators. RPMs, such as Sm-Co and Nd-Fe-B, have 8
to 15 times the energy product compared to the ferrites. The
high-energy product should significantly benefit the inno-
vative design with tremendous performance enhancement.
However, reports of multi-megawatt wind turbine showed
similar generator configurations, even when these very dif-
ferent magnets ferrites and Nd-Fe-B would be used.[39] For
example, a 144-pole radial-flux PM generator would use 2948 kg
of ferrite and a 144-pole radial-flux PM generator would use
Figure 10. Clipper’s 2.5 MW wind turbine with four permanent magnet
generators. Reproduced with permission.[41]
830 wileyonlinelibrary.com © 2011 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
www.MaterialsViews.com
2918 kg of Nd-Fe-B. Apparently these proposed generators do
not fully utilize the high energy density of RPMs and therefore
a new design of PM generators is needed.
A 3 MW DD generator uses about 1.5 tons of RPMs; with
this the magnetic material represents a significant fraction of
the generator costs. Due to the consumer demand for low cost
and long lifespan wind turbines, the permanent magnet com-
ponents represent an important trade-off between minimizing
volume and protection against irreversible demagnetization due
to electrical faults or excessive heating. As demand increases
for wind turbines and permanent magnet technologies emerge
across industry sectors, a high demand for rare-earth magnets
could result in an increased material price and consequent chal-
lenges for the economical feasibility.
2.5. Energy Efficient Electric Machines: Electric and Hybrid
Vehicles
The year 2009 was the 100th anniversary of the granting of the
first US hybrid car patent or “Mixed Drive for Autovehicles”
(Figure 11) as it was described by the German inventor Henri
Pieper.[42] He says in his patent: “The invention… comprises
an internal combustion or similar engine, a dynamo motor
direct connected therewith, and a storage battery or accumu-
lator in circuit with the dynamo motor, these elements being
cooperatively related so that the dynamo motor may be run as
a motor by the electrical energy stored in the accumulator to
start the engine or to furnish a portion of the power delivered
by the set, or may be run as a generator by the engine, when
the power of the latter is in excess of that demanded of the set,
and caused to store energy in the accumulator.” Pieper’s patent
also included a rudimentary battery and engine management
system. Unfortunately for Pieper, the year before his patent
was granted the assembly lines for the first inexpensive mass-
produced car based on the internal combustion engine, the Ford
Model T, were under construction. By the 1930s, production of
hybrids based on the Pieper system had ceased completely and
the prevalence of the gasoline engine powered automobiles was
cemented. Drivers controlled a magnetic-disc clutch that con-
nected the engine to a DC-motor generator, or dynamo, which
was in turn connected to a gear-set that turned the rear wheels
via chain drive. Now, a century later, hybrid electric vehicles
have re-emerged as a realistic alternative to the gas-only internal
combustion vehicle. The basic concept of a parallel hybrid is in
principle retained in today’s hybrids like the Toyota Prius (see
Figure 12).
Currently more than 1.5 million HEVs are on the road. Apart
from the worldwide environmental issues, this re-emergence
has been driven by the multiple advances in magnetic mate-
rials, electrical machine design, power electronics, and battery/
fuel cell technology. Electric or hybrid vehicles are being devel-
oped at an unprecedented pace[44] and their sales is expected
to increase dramatically in the next five years. One of the three
key components (besides the power control unit and battery) of
the hybrid vehicles is the PM-electric motor. As described ear-
lier, as is the case with PM generators, PM motors have many
advantages compared to induction motors including compact
size, high efficiency, light weight, and high torque. According
Adv. Mater. 2011, 23, 821–842

www.MaterialsViews.com
Figure 11. Picture from Henri Pieper’s hybrid vehicle patent application
from 1905.[42]
to Toyota, each hybrid Prius motor/generator uses ≈1.3 kg of
Nd-Dy-Fe-B magnets.[43] The Toyota Hybrid System (THS) is a
power train combining the series hybrid system with the par-
allel hybrid system in order to maximize the benefits of both.
Developing energy-efficient machines and hybrid vehicles
demands permanent magnets with high performance as shown
in Figure 12.
One way to reach the goal of having the best fuel economy
and acceleration in the marketplace is to develop new permanent
magnets with higher energy product and good thermal stability
at temperatures of 150–250 °C. Current hybrid vehicles use Nd-
Dy-Fe-B magnets with (BH)max = 286.5 kJ m−3 (36 MGOe) and
intrinsic coercivity iHc = 2400 kA m−1 (30 kOe), stable at tem-
peratures of about 180 °C. The value of iHc is very important
for such dynamic applications with elevated operating tempera-
ture because it is a key to resisting thermal demagnetization.
The performance of a PM motor is directly related to the mag-
netic flux provided by the permanent magnets. The Prius PM
motor consists of an eight-pole rotor with high coercivity Nd-
Fe-Dy-B magnets inside. As shown in Figure 13, by arranging
the permanent magnets in a V-shape rather than a flat-shape,
the output power is increased from 33 to 50 kW (≈1.5 times)
and the drive torque is increased from 350 to 400 N-m (≈15%
Figure 12. Prius’ generator and motor. Reproduced with permission from Toyota Motor Corporation.[43]
Adv. Mater. 2011, 23, 821–842 © 2011 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
www.advmat.de
REVIEW
higher). The increased performance results from higher mag-
netic flux density in the working gap, estimated at ≈15% using
Ansoft’s Maxwell 3D computer modeling.[45]
It is not a straight-forward calculation to determine the per-
cent gain in motor efficiency resulting from the gain in energy
density of the magnet (BH)max. Different design is required as
the energy density of the magnet is increased because a higher
magnetic flux from the magnets can cause localized magnetic
saturation in certain areas if the design is not changed, and
this will not increase the torque and output power of the motor.
Many researchers reported their attempts to establish a method
to estimate the torque of a motor.[46] These reports targeted dif-
ferent PM-rotor design rather than different energy densities of
the magnets. To estimate the motor’s improvement resulting
from the increase in energy density of the magnets, Ansoft’s
RMxprt computer software was used for analyzing a four-pole
surface-mounted PM motor,[45] as shown in Figure 14, using
the magnets with energy products of 23, 36, and a fictitious
72 MGOe. The results are shown in Figure 15. When the same
motor configuration is used, the starting torque (rpm = 0) and
output power are increased by 25−30% when (BH)max of the
magnet is increased from 36 to 72 MGOe. However, we have
to realize that a new design must be used when better mag-
nets are used, as described earlier. According to our estimation,
the torque and output power with 72 MGOe magnets could be
increase by 70% compared to 36 MGOe magnets if an appro-
priate design were be used. The torque and the output power of
the new design using 72 MGOe magnets are shown in the top
plots in Figure 15.
2.6. Understanding and Controlling Microstructure
on the Nanoscale
Various approaches are being examined in order to develop
Dy-free Nd-Fe-B sintered magnets with high Hc for
high-temperature applications.[47] In order to achieve this, it is
wileyonlinelibrary.com 831
 www.advmat.de
REVIEW
Figure 13. The effect of magnet arrangement on the performance of Prius PM motor/gen-
[ 43 ]
erator : magnet arrangements in 8-pole rotors of THS motor (top left) and THS II motor (top
right) and related torque (bottom left) and output (bottom right) for the two arrangements. and soft magnetic nanoparticles. Figure 17a,c
Reproduced with permission from Toyota Motor Corporation.[43]
vital to understand the coercivity mechanism in Nd-Fe-B sin-
tered magnets in greater detail. The Nd-rich phases and their
distributions in the microstructure are critical for the Hc of the
magnet, but currently Hc reaches only 20–30%[8] of the theo-
retical maximum (Brown’s paradox[48]), which is the anisotropy
field, HA. This is because the nucleation of inverse magnetic
domains occurs at a much lower magnetic field than the anisot-
ropy field starting from the grain boundaries, where the anisot-
ropy of the Nd2Fe14B phase is expected to be locally decreased
due to the presence of defects, roughness, or incomplete wet-
ting by the paramagnetic Nd-rich grain boundary phase. Since
the coercivity of Nd-Fe-B magnets is strongly influenced by the
microstructure, especially the magnetic isolation of the ferro-
magnetic grains and the grain boundary structure and micro-
chemistry, important microstructure changes are expected to
occur during the postsinter annealing, particularly at the grain
boundaries.[49] However, there is no definite work that has char-
acterized the structure and chemistry of the grain boundaries
of Nd-Fe-B sintered magnets. Magnetically isolated single-
domain particles are expected to give rise to much higher
Figure 14. The 2D and 3D model of a four-pole surface PM motor used
for computer modeling (the copper windings are hidden).[21]
832 wileyonlinelibrary.com © 2011 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
www.MaterialsViews.com
coercivity, but such a microstructure has
never been achieved in the Nd-Fe-B system.
The approach of multiscale characterization
of sintered Nd-Fe-B commercial magnets is
illustrated in Figure 16.
On another front, research in producing
exchange-coupled nanocomposite perma-
nent magnets has achieved remarkable
progress.[50] By adding nanoscale soft mag-
netic phases into a hard magnetic phase
matrix, the energy density of the composite
can be greatly enhanced due to the inter-
phase exchange coupling,[51] while the cost of
the magnets can be simultaneously reduced
by the lower rare-earth contents of the com-
posites compared to their single-phase coun-
terparts. A bottom-up approach has been
applied to produce nanocomposite magnets
based on rare-earth transition metal com-
pounds starting with preparation of the hard
show the nanoscale particles produced by
surfactant-assisted ball milling.[52,53] The par-
ticles are magnetically anisotropic and can be
aligned in a magnetic field to form anisotropic bulk magnets.
Figure 17d–f show the Fe-Co-based soft-phase distribution in a
bulk SmCo5/Fe(Co) nanocomposite processed by severe plastic
deformation and warm compaction.[54] In the isotropic bulk
magnets processed using these techniques, substantial energy
density enhancement has been achieved while ≈30% fewer
rare-earth elements were used.[54] It is predicted that success
in producing anisotropic nanocomposite magnets will lead to
a revolutionary change in permanent magnetic materials devel-
opment and applications.
3. Soft Magnetic Materials
Soft magnetic alloys have played a key role in both power gen-
eration and conversion for the electrical grid, as one can see
from the heavy application of soft magnets in transformers.
The need for efficient generation, transmission, and distribu-
tion of electric power is ever growing; however, at the same
time the annual electric losses are outpacing annual increases
in electricity consumption. While the increased production of
energy from renewable sources is extremely important, the
reduction of electrical losses also needs to be addressed. At
present, the electrical grid consists of localized stations for the
generation of electricity. In the USA, electricity is converted to
high-voltage AC current at voltages between 138 and 765 kV and
transmitted to substations near its end-use location. The voltage
is then stepped down to lower values (between 13 kV and 120 V)
for distribution to industrial and residential consumers. These
generation, transmission, and distribution systems are aging,
inefficient, and inadequate to meeting the future energy needs
of the USA without significant changes in operation and infra-
structure. For these reasons, advanced electric storage systems,
smart controls, and power electronics for AC–DC conversion are
technologies that are being supported to revolutionize the way
Adv. Mater. 2011, 23, 821–842
 www.advmat.de
www.MaterialsViews.com

REVIEW
Figure 15. The effect of the energy products of the magnets on a PM motor performance (all magnets have linear B–H demagnetization curves in the
second quadrant suitable for dynamic application). Torque vs speed (left) and output power vs speed (right).[45]

we deliver power.[55] These so-called “smart-grid” technologies, desire to increase the capability of these devices, induction gen-
using an array of advanced materials and devices, will provide erators housed in the nacelle must be given higher energy den-
more efficient, affordable, and sustainable energy use for the sities, which can be accomplished with improved soft magnetic
long term.[56] On the electricity generation side, soft magnetic materials and/or permanent magnetic materials if PM-motor
materials are especially important to maintaining an efficient designs are employed.
smart grid. For example, soft magnets have important roles in About 9% of the electricity generated is lost as heat during
wind energy technologies for generation using advanced induc- transmission and distribution.[59] While high voltage is neces-
tion motors and for conversion using power converters and/or sary for low losses in electrical transmission, DC transmission
power transformers. The architecture of modern wind turbines has lower losses than AC transmission. This is one reason that
include the generator in the nacelle (the enclosure attached to the smart grid has been discussed in terms of high-voltage
the rotor blades at the top of the turbine tower).[57,58] Due to the direct-current (HVDC) transmission. The expense of the HVDC
converter stations limits the implementation of this
more energy-efficient technology to transmission.
For this reason, the implementation of smart grid
technology will continue to require improved energy
efficiency for rectifiers, inverters, and transformers
using advanced soft magnetic materials. In present
devices, losses can be as low as 0.25% for some large
utility transformers, however, they may also be as
high as 15% for small consumer electronics “power
bricks”. In these applications, losses come from
a variety of sources including winding resistance,
eddy currents, stray fields, hysteresis, mechanical
losses, magnetostriction, and cooling systems. Of
these, eddy currents, hysteresis, and magnetostric-
tion are important soft magnetic parameters. Due
to their central role in these devices, soft magnetic
cores with lower core losses (including contributions
from hysteretic, magnetostrictive, and eddy current
Figure 16. Multiscale characterization of sintered Nd-Fe-B commercial magnets:
sources), higher magnetization, and improved cost
a) Backscattered electron contrast scanning electron microscopy (SEM) image effectiveness will be key components for improved
showing the hard magnetic Nd2Fe14B matrix grain well-separated by a thin Nd-rich energy efficiency.
intergranular phase (white contrast). b) Electron backscattering diffraction (EBSD) Soft magnetic materials come in many forms
image quantitatively showing the local variation in orientation of the different grains. ranging from soft iron, used in powder core induc-
c) optical Kerr microscopy showing the corresponding magnetic domain structure of tors, to silicon steel laminates and nanocrystalline
the same area. d) Transmission electron microscopy (TEM) image showing the amor-
or amorphous tape wound cores for transformers,
phous Nd-rich intergranular phase of about 1 nm thickness (from top left to bottom
right). Panels (e,f) show schematically and in atomic resolution the Nd2Fe14B matrix and to soft ferrites for electromagnetic interference
phase. Courtesy: T. G. Woodcock, IFW Dresden. (EMI) filters. The materials selection is important for

Adv. Mater. 2011, 23, 821–842 © 2011 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim wileyonlinelibrary.com 833
 www.advmat.de
www.MaterialsViews.com
REVIEW

frequencies. These advanced classes of alloys provide

flexibility in device design.
Most transformers are designed for optimal
energy efficiency, splitting the losses of the system
between those generated by the core and those gen-
erated by the copper windings (Figure 19). This is
done to transfer the maximum amount of power
per core volume and to keep the rise in the tempera-
ture of the core to a minimum. The resulting design
shows that the lowest total losses are coincident with
an optimal flux density change for the transformer.
The position of the optimal flux density is dependent
on the core material and number of turns on the
transformer. The saturation of the magnetization of
the core material provides an upper limit to the flux
density change (although in practice the maximum
useful flux density is considerably lower due to
reduced permeability of the core as a near saturating
Figure 17. Nanoscale hard magnetic particles and composites. a) TEM image of ≈6 nm field is applied).
particles of Sm2Co17 produced by surfactant-assisted ball milling; b) TEM image of While utility transformers tend to have high effi-
≈20 nm particles prepared in the same method with size selection; and c) SEM image ciency (above ≈97%), their ubiquitous use means that
of the SmCo5 plate-like particles obtained by surfactant-assisted ball milling. d) Energy- even small improvements in core loss can have large
filtered TEM image showing Fe distribution and e) Sm distribution in a SmCo5/Fe energy savings systemically. Electrical steels (e.g.,
nanocomposite bulk sample with high Fe fraction of ≈30%. f) The 3D atom probe
Fe 3%Si) are typically used in utility transformers
analysis for the SmCo5+ Fe nanocomposite bulk magnet with 60% Fe isoconcentration
surface. Panel (b) adapted with permission.[53] Copyright 2007, Institute of Physics. due to their low cost and high magnetization. How-
ever, their core losses are substantially higher than
recently developed nanocrystalline materials.[65] The
each application to provide the best balance of performance and move to high-energy-efficiency magnetic cores requires a larger
cost. While the application dictates which performance met- initial capital investment however a significant reduction in
rics dominate, in general good soft magnets have the following no-load losses, almost entirely due to the magnetic material,
properties: high saturation magnetization, resistivity, and per- provides a cost savings over the lifetime of the transformer due
meability, and low coercivity and core losses. Operation at ele- to the improved efficiency.[66] Higher efficiency also results in
vated temperatures requires a high Curie temperature. Of these reduction of CO2 emission since less electrical energy is wasted
parameters, the ones most closely tracked for application are in the transmission and distribution process.[67] Conversion
core losses (as a function of maximum applied induction and technologies for wind, solar, and storage by flywheel technology
frequency) and saturation magnetization. Representative values (i.e., off-peak energy storage of electricity generated by wind
for select soft magnetic alloys are shown in Table 6. Graphically, and solar technologies) will also require magnetic converters
the classes of soft magnetic materials are defined in Figure 3 to change the solar or wind directly (or the stored mechanical
with the bounds of saturation magnetization and coercivity energy in the flywheel) to electrical energy.[68]
values shown. Improvements in transportation technologies will also
Since the origin of core losses is strongly dependent on the demand advanced soft magnetic materials. As hybrid electric
choice of material, the materials selection can result in very dif- and battery electric vehicles are developed and produced in
ferent optimal component designs. Figure 18 shows the core increasing numbers, the need for miniaturized, high efficiency
losses for selected soft magnetic materials as a function of the advanced induction motors and power converters to charge
maximum flux density and using a sinusoidal switching field at these vehicles will require advances in materials. In this case,
50 or 60 Hz. The amorphous (e.g., Metglas) and nanocrystalline large magnetization and higher frequency operation will allow
(e.g., Finemet) alloys provide the greatest energy efficiency with miniaturization of the motors and generators, providing better
the lowest core losses for a given applied magnetic field (or flux efficiency for the system. Elevated temperature power genera-
density) and frequency (not shown). This is largely due to their tion and conversion will be necessary for electric aircraft and
low values of magnetocrystalline anisotropy, which determines ships, as well as improved efficiency at higher operation fre-
the resistance of domain wall motion through the material quencies for power electronics devices. Switch-mode power
when a magnetic field is applied. The coercivity is a parameter supplies used in power electronics applications require mate-
from Table 6 that reflects the low value of magnetocrystalline rials with a square magnetic hysteresis loop and low AC losses
anisotropy and is proportional to the hysteretic losses in the for magnetic amplification purposes as well as a broadband
material (a component of the total core losses). Amorphous high permeability core to smooth out the ripple current.[69] As
and nanocrystalline alloys have the added benefit of moderate distributed power sources become more widespread, the neces-
resistivity, which limits the effects of eddy currents on the sity for DC/DC conversion of large output currents will require
total core losses allowing them to be used at higher switching advanced core materials to operate efficiently at low voltages

834 wileyonlinelibrary.com © 2011 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim Adv. Mater. 2011, 23, 821–842
 www.advmat.de
www.MaterialsViews.com

Table 6. Magnetic properties for various commonly used soft magnetic materials.[60–62] Note: the Metglas alloy is amorphous and the Finemet

REVIEW
alloy is nanocrystalline. One Tesla exceeds the saturation magnetization for alloys marked with †.

Material μ0Ms (T) μmax (103) Hc (A/m) electrical resistivity Tc (°C) Pcm [60 Hz/1 T] magnetostriction
(ρe) (μΩcm) (W/kg) (λs) (ppm)
Fe49Co49V2 2.4 5–50 16–398 27 930 1.10 60
Soft Fe 2.16 10–50 4–80 10 771 20 −2
Fe 3%Si (GO) 2.0 0.5–5 6 48 745 0.84 7
Fe67Co18B14Si1 1.8 50 3.5 123 415 0.66 35
Metglas 2605CO
Ni50Fe50 permalloy 1.4–1.6 70 4–20 40–50 480 0.33 18
Fe73.5Si13.5Nb3B9Cu1 1.23–1.35 80 0.6–2.5 110 570 0.01 0–2
Finemet
Ni78Fe17Mo5 0.65–0.82 100–800 0.25–0.64 60 400 † 2–3
supermalloy
Mn-Zn ferrite 0.36–0.5 0.5–10 10–100 107–108 150–250 † ±5
Ni-Zn ferrite 0.25–0.42 0.01–1 14–1600 1011 120–500 † −20

and high frequencies (in conjunction with a move to GaAs or
SiC semiconductor active components).
Size and weight reductions can be achieved for a given output
power requirement by increasing the operation frequency, as
demonstrated in the schematic diagram from Figure 20.[70,71]
For instance, for a 500 W output power the volume of the
transformer can be reduced by 50% when the frequency is
raised from 6 to 60 kHz. As size and weight are extremely
important for transportation applications (including electric
and hybrid cars and ships), exploitation of this effect would be
very advantageous, however, the losses of the system are not
represented in this figure, and both the core and copper losses
increase with increasing frequency. Recent improvement in
soft magnetic performance by microstructure and composition
control have been active areas of research to provide more
material options for better energy efficiency at higher operation
frequencies.[72,73]
Great improvements in energy efficiency have been demon-
strated for amorphous and nanocrystalline soft magnetic mate-
rials. Due to the higher capital costs, this technology has not
received wide-spread use despite the availability of promising
materials research during the past 20 years, and the lower life-
cycle cost of using these cores.[66] The use of amorphous alloys
has been shown to reduce transformer size and reduce acoustic
noise when compared to conventional Si-steels used in most
transformer cores today. Additionally, the improved energy effi-
ciency of transformers using these materials results in reduced
carbon emissions considering the majority of electricity is gen-
erated from fossil fuels.
While recent soft magnetic materials developments have
resulted in advances in low-loss magnetic alloys, many of
these developments require sustained research investment to
improve the magnetization, magnetostriction, and extend their
upper operation frequency. Investment in the development of

Figure 19. Schematic diagram showing the contributions of copper and

Figure 18. Core loss as a function of maximum flux density for various core losses used in transformer design with the most energy efficient
soft magnetic materials with a 50 or 60 Hz switching field.[62–64] design indicated as optimum.

Adv. Mater. 2011, 23, 821–842 © 2011 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim wileyonlinelibrary.com 835
 www.advmat.de
REVIEW
Figure 20. Schematic diagram showing the relationship between required
power output, switching frequency, and volume as a function of optimum
changes in flux density for a model switchmode transformer.
high-efficiency soft magnetic materials now will enable more
flexible engineering designs for electronics and transportation
to meet tomorrow’s energy needs.
4. Magnetocaloric Materials
Modern society relies on readily available refrigeration for
protecting food and providing comfortable living spaces. Con-
ventional refrigerators use ozone depleting chemicals (chlo-
rofluorocarbons (CFCs)), hazardous chemicals (ammonia
(NH3)), or greenhouse gases (hydrochlorofluorocarbons
(HCFCs) and hydrofluorocarbons (HFCs)) in a vapor-
compression cycle to provide cooling. While they have met the
cooling needs and improved the standard of living for people
worldwide, conventional refrigerators tend to be unwieldy,
hefty, and lack energy efficiency. Recently, an alternative refrig-
eration method using adiabatic demagnetization (the so-called
magnetocaloric effect (MCE)) has been investigated as a means
to address the short-comings of vapor-compression refrigera-
tion. Magnetic refrigeration has three prominent advantages
compared to compressor-based refrigeration. First there are no
836 wileyonlinelibrary.com © 2011 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
www.MaterialsViews.com
harmful gasses involved, second it may be built more compactly
because the main working material is a solid, and third mag-
netic refrigerators generate much less noise. Also, the cooling
efficiency in magnetic refrigerators working with gadolinium
has been shown to reach 60% of the theoretical limit,[74] com-
pared to only about 45% in the best gas-compression refrigera-
tors. While commercial refrigerators of this type are still in the
development stages, research efforts to develop new materials
with improved MCE are focused on maximizing the cooling
capacity and energy efficiency of this budding technology. In
this section, the different materials are compared, concen-
trating on transition metal containing compounds.
When a material is magnetized in an applied magnetic field,
the entropy associated with the magnetic degrees of freedom,
the so-called magnetic entropy, Sm, is changed because the field
changes the magnetic order of the material. Under adiabatic
conditions, ΔSm must be compensated by an equal but opposite
change in the entropy associated with the lattice, resulting in
a change in the temperature of the material. This temperature
change, ΔTad, is usually called the MCE. It is related to the mag-
netic properties of the material
  through the thermodynamic
Maxwell relation ∂∂ BS T = ∂∂M T B
. From magnetization meas-
urements made at discrete temperature intervals, ΔSm can be
calculated as described in Refs. [75,76]. For materials showing a
first-order phase transition with large hysteresis these magneti-
zation measurements need to be done with great care to avoid
overestimation of the entropy change.[77] On the other hand, the
magnetic entropy change can be obtained more directly from
a calorimetric measurement of the field dependence of the
heat capacity, c, and subsequent integration. It has been con-
firmed that the values of ΔSm(T,B) derived from the magnetiza-
tion measurement coincide with the values from calorimetric
measurement.[78] The adiabatic temperature change, ΔTad(T,B),
can then be integrated numerically using the experimentally
measured or theoretically predicted magnetization and heat
 
capacity. Obviously, the MCE is large when ∂∂M T B
is large and
c(T,B) is small at the same temperature. As we are interested
in effects at higher temperatures, the heat capacity is gener-
ally quite large, on the order of the Dulong–Petit law, c ≈ 3NR,
where N is the number of atoms and R is the molar gas con-
stant. Therefore, we should concentrate on finding a large
change
 ∂ M  in magnetization at the relevant temperature. Since
∂ T B peaks at the magnetic-ordering temperature, a large
MCE is expected close to this magnetic phase transition and the
effect may be further maximized, when the order parameter of
the phase transition changes strongly within a narrow tempera-
ture interval. The latter is true for first-order phase transitions.
In the magnetic-refrigeration cycle, depicted in Figure 21,[79,80]
initially randomly oriented magnetic moments are aligned by a
magnetic field, resulting in heating of the magnetocaloric mate-
rial. This heat is removed from the material to the ambient envi-
ronment by heat transfer. Upon removing the field, the magnetic
moments randomize, which leads to cooling of the material below
ambient temperature. Heat from the system to be cooled can then
be extracted using a heat-transfer medium. Depending on the
operating temperature, the heat-transfer medium may be water
(with antifreeze) or air and, for very low temperatures, helium.
Therefore, magnetic refrigeration is an environmentally friendly
cooling technology. Current research aims at new magnetic
Adv. Mater. 2011, 23, 821–842

www.MaterialsViews.com
materials with larger magnetocaloric effects, which then can be
operated in fields of about 2 T or less (i.e., fields that can be gen-
erated by the aforementioned Nd-Fe-B-type permanent magnets).
4.1. La(Fe,Si)13 and Related Compounds
The Fe-rich compounds La(Fe,Si)13, with the cubic NaZn13-type
structure, possess magnetic ordering temperatures between
200 and 262 K and exhibit typical invar-type behavior.[81,82]
Due to the sharp first-order magnetic-ordering transition,
several groups have reported a large magnetocaloric effect
for this class of compounds,[83,84] which is enhanced when
the materials are processed by rapid quenching and subse-
quent annealing.[85–87] Above the Curie temperature Tc, an
itinerant-electron metamagnetic (IEM) transition is induced by
an external magnetic field. The IEM transition occurs due to
the change in the density of states at the Fermi level under the
action of the magnetic field. Both transitions to the ferromag-
netic state are accompanied by a sizable volume change of about
1%.[88] In real materials, the lattice parameter can be modified
physically by applying external pressure or chemically by the
addition of substitutional or interstitial alloying elements.[89]
To increase the magnetic-ordering temperature without signifi-
cant loss in the magnetic moment, one may replace some Fe by
other magnetic transition metals. The compounds La(Fe,Co)13–
xAlx and La(Fe,Co)13–xSix with x ≈ 1.1 and therefore a very high
transition metal content show a considerable magnetocaloric
effect near room temperature.[90,91] This is achieved with only
a few percent of Co and the Co content can easily be varied to
tune the critical temperature to the desired value, however, near
room temperature the values for the entropy drop.
Further improvement in the magnetocaloric effect can be
achieved by expanding the crystalline lattice, thereby influ-
encing the character of the direct exchange interaction (from
Figure 21. Schematic representation of a magnetic-refrigeration cycle that
transports heat from the heat load to the ambient environment. Yellow
and green depict materials in low and high magnetic fields, respectively.
Reproduced with permission.[80] Copyright 2005, Institute of Physics.
Adv. Mater. 2011, 23, 821–842 © 2011 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
www.advmat.de
REVIEW
antiferromagnetic to ferromagnetic). The lattice expansion due
to the addition of three interstitial hydrogen atoms per formula
unit is about 1.5%.[92–94] The critical temperature can be increased
to up to 450 K, the average magnetic moment per Fe increases
from 2.0 μB to up to 2.2 μB and the field- or temperature-induced
phase transition is found to be of first-order for all hydrogen
concentrations. This all results for a certain Si percentage in an
almost constant value of the magnetic entropy change per mass
unit over a broad temperature range (Figure 22).
Material processing will be somewhat more elaborate than
for simple metal alloys, however La is the cheapest of the rare-
earth elements so these materials should be cost-effective. Due
to the above described changes in lattice parameters caused
by the phase transitions and/or hydrogenation, the material
will potentially have long-term cycling problems because the
polycrystalline material may become granular with use. This
can have distinct influence on the corrosion resistance of the
material and thus on the lifetime of a refrigerator and therefore
must be investigated. Very recently, it could be demonstrated
that powderized bulk materials and melt-spun particles can be
hot-compacted to a porous morphology. The porous structure is
much more tolerant to the huge magnetoelastic effects related
to the first-order-type transitions and retains their mechanical
integrity over hundreds of cycles.[96] Only a slight decrease in
the adiabatic temperature change compared to the conventional
bulk material was observed and, with this, one of the major
obstacles for application can be regarded as being overcome.
4.2. MnAs-Based Compounds
MnAs exists in two distinct crystallographic structures.[97] At
low- and high-temperature the hexagonal NiAs structure is
found and for a narrow temperature range from 307 to 393 K
the orthorhombic MnP structure exists. The high-temperature
transition in the paramagnetic region is of second-order.
The low-temperature transition is a combined structural and
ferro-paramagnetic transition of first-order with large thermal
20
Gd metal
LaFe11.4Si1.6H
LaFe11.4Si1.6H1.5
15
LaFe11.4Co0.5Si1.1
LaFe11.2Co0.7Si1.1
-ΔS (J/kg K)
10
0
260 280 300 320 340
T (K)
Figure 22. Magnetic entropy change for different LaFe13-based samples
at a field change of 2 T.[92,95]
wileyonlinelibrary.com 837
 www.advmat.de
REVIEW
hysteresis. The change in volume at this transition amounts
to 2.2%.[98] The transition from paramagnetic to ferromagnetic
occurs at 307 K, the reverse transition from ferromagnetic to
paramagnetic occurs at 317 K. Very large magnetic entropy
changes are observed in this transition.[99] In the concentra-
tion range from 5 to 40% Sb, TC can be tuned between 220 and
320 K without losing much of the magnetic entropy change.[100]
Direct measurements of the temperature change confirm a
ΔT of 2 K T−1.[101] When Mn is substituted by Cr the thermal
hysteresis is reduced and Tc is lowered, while the large entropy
change is maintained.[102]
The materials costs of MnAs are quite low but processing of
As containing alloys is complicated due to the biological activity
of As. In the MnAs alloy the As is covalently bound to the Mn
and would not be easily released into the environment. However,
this should be experimentally verified, in particular because in
an alloy second phases frequently form and may be less stable.
The change in volume in MnAs and MnSb is still 0.7% which
may result in aging after frequent cycling of the material.
4.3. Heusler Alloys
Heusler alloys frequently undergo a diffusionless, displacive or
martensitic transformation between austenite, a high temperature,
cubic, parent phase, and martensite, a low temperature, tetragonal
phase. The transformation is generally temperature-induced and
of first-order. While this effect has been carefully considered for
magnetic shape memory applications,[103] it may also be exploited
for magnetic refrigeration. Very close to the martensitic-transition
temperature, one observes a large change in magnetization for
low applied magnetic fields due to the change in magnetocrystal-
line anisotropy between the austenite and martensite phases. This
change in magnetization can result in a moderate MCE.
An example of a material from this class with good magneto-
caloric response is Ni2MnGa, which has a ferromagnetic Curie
temperature of 376 K and a martensitic-transformation tem-
perature near 220 K.[104] This martensitic-transformation tem-
perature can be easily changed to around room temperature
by modifying the composition of the alloy from the stoichio-
metric one. As may be expected from its cubic structure, the
parent phase has a low magnetocrystalline anisotropy (μ0Ha =
0.15 T). However, the martensitic phase exhibits a much larger
anisotropy (μ0Ha = 0.8 T). The corresponding changes in the
response to an applied magnetic field during the transforma-
tion results in a moderate magnetic entropy change of a few
J mol−1 K−1 (Figure 23), which is enhanced when measured on
a single crystal.[105,106] When the composition in this material is
tuned in a way that the magnetic and structural transformation
occurs at the same temperature, the largest magnetic-entropy
changes are observed.[107]
Some of Heusler alloys with In or Sn as the p-metal show
an inverse MCE. These materials cool down upon application
of a magnetic field.[110–112] This effect is related to the differ-
ence in exchange interaction for the austenitic and martensitic
phase and may be useful for special applications. A detrimental
feature of the martensitic transition, the rather large thermal
hysteresis, can to some extend be controlled by either compo-
sition, processing,[113] or hydrostatic pressure.[114] Additionally,
838 wileyonlinelibrary.com © 2011 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
www.MaterialsViews.com
Gd
NiMnGa
30
NiMnGa
MnAs
MnAs0.9Sb0.1
MnAs0.97Si0.03
-ΔS (J/K kg)
20
10
0
260 280 300 320 340 360
T (K)
Figure 23. Magnetic-entropy change for MnAs and Mn1+δ As0.9Sb0.1,[100]
MnAs0.97Si0.03[108] and two NiMnGa[109] alloys at a field change of 2 T.
the extremely large length changes related to the martensitic
transition will definitely result in aging effects. It is well-known
for the magnetic shape memory (MSM) alloys that only single
crystals can be frequently cycled and polycrystalline materials
spontaneously powderize after several cycles. The synthesis
of polymer-composite MSM materials does circumvent this
effect.[115,116]
4.4. Fe2P-Based Compounds
Another phenomenon leading to a large MCE is the discon-
tinuous volume change that accompanies the magnetic transi-
tion observed in Fe2P-type alloys with Mn and As substitutions.
The Curie temperature of this compound is 216 K and the
magnetic transition is of first-order.[117] The magnetic-ordering
transition from the paramagnetic state to the ferromagnetic
state is accompanied by a discontinuous change in the volume
of 0.05%. Thus, the ferromagnetic state has a higher volume
than the paramagnetic one. This phase transition is found to
be extremely sensitive to changes in pressure or magnetic field.
Substitution of As, B, or Si into the P sublattice results in an
increase of the Curie temperature, which can easily be lifted to
above room temperature for As or Si concentrations of 10% or
B concentrations of 4%.[118]
The most extensively studied series of alloys is of the type
MnFe, MnP, and MnAs. The most striking feature of the
MnFeP–MnFeAs phase diagram is the stability of the hexag-
onal Fe2P-type structure for As concentrations between 30 and
65%.[119] This phase has ferromagnetic order accompanied by a
discontinuous change in volume and a resultant large magne-
tocaloric effect.[120–122] While the total magnetic moment is not
affected by changes in the composition, the Curie temperature
can be tuned from 200 to 450 K.
For these compounds one observes a strong change in the
critical temperature on the application of a magnetic field.
This change in critical temperature is a good indication for the
temperature change induced by the application of a magnetic
field.[123] The thermal hysteresis is signature of a first-order
Adv. Mater. 2011, 23, 821–842
 www.advmat.de
www.MaterialsViews.com

REVIEW
phase transition. Because of the small size of the thermal hys- 40
teresis (less than 1 K), the magnetization process can be consid- Gdmetal
ered to be reversible in temperature. Variation of the Mn:Fe ratio MnFeP 0.45As0.55
may also be used to further improve the MCE. More recently, a MnFeP 0.5As0.3Si0.2
30
large MCE was observed in the compound MnFeP0.5As0.3Si0.2 at MnFeP 0.5Si0.3Ge0.2
room temperature (see Figure 24).[124] After replacing all As, a
MnFeP 0.5As0.5

-ΔS (J/K kg)

considerable large MCE is still observed for MnFe, MnP, MnSi,
and MnGe.[125,126] However, these alloys exhibit a much larger
thermal hysteresis. Processing and variation of the stoichiom-
etry on the transition-metal sublattice appear to help avoid the
undesired hysteresis.[127,128]
The excellent magnetocaloric features of the compounds of
the type MnFe, MnP, MnSi, MnGe, and MnAs and the very
low material costs make them attractive candidate materials for
commercial magnetic refrigerators. However, as is the case for
MnAs alloys it should be verified that materials containing As
do not release this to the environment. The fact that the magne-
toelastic phase transition is a change of the lattice parameters
rather than a change in volume makes it feasible that this alloy,
even in polycrystalline form, will not experience severe aging
effects after frequent magnetic cycling.
4.5. Comparison of Different Materials and Outlook
The MCEs for field changes of 2 T are summarized in
Figure 25. It is obvious that above room temperature a few
transition-metal-based alloys perform the best. If one takes into
account the fact that ΔT also depends on the specific heat of
the compound[129] these alloys are still favorable and not only
from the cost point-of-view. This makes them likely candidates
for use as magnetic refrigerant materials above room tempera-
ture. However, below room temperature a number of rare-earth
compounds perform better and for these materials a thorough
cost versus performance analysis will be needed.
The main parameters of the various materials are also sum-
marized in Table 7, permitting a fast comparison. At present
it is not clear which material will get to the stage of real life
applications. Though it is already feasible that for applications
with limited temperature span and a cooling power in the kW
range like air conditioning, commercial competitive magnetic
refrigerators are quite possible, it is not yet obvious, which of
the above mentioned materials shall be employed. A variety
of prototypes have been built[130,131] but materials are not yet
available at large scale. Though most attention is paid to the
MnFeP 0.8Ge0.2
20
10
0
260 280 300 320
T (K)
Figure 24. Magnetic-entropy changes of Fe2P type compounds at mag-
netic field changes of 2 T.[120–122]
pure magnetocaloric properties and materials costs, recently
also other properties like dynamics of the phase transition,[132]
corrosion resistance, mechanical properties, heat conductivity,
electrical resistivity and environmental impact are getting atten-
tion. With the refrigeration market being a multibillion dollar
market, this novel technology offers great opportunities.
Taking into account the different requirements for magnetic
refrigerants, it may be stated that the ideal magnetic refrigerant
should at least contain 80% transition metals with a large mag-
netic moment such as Fe or Mn. In addition it should contain
some inexpensive p-metal such as Al or Si, which can be used
to tune the working point of the material. This material should
then exhibit a magnetic ordering transition of first-order and
be as workable as steel, with a corrosion resistance similar to
stainless steel and a high electrical resistance.
5. Summary
In this review we have described an existing but somehow
largely ignored fact that improved magnetic materials are a
solution to the energy crisis. The solution via development in
magnetic materials will give an immediate impact in producing
renewable energy and in raising energy efficiency. Magnetic
materials have already played a significant role in energy

Table 7. Comparison of different potential magnetocaloric materials. Gd is included as reference material ([d] means direct measurement, [c] is calcu-
lated from a combination of measurements). The costs may strongly fluctuate due to market demands and quality of starting materials. Reproduced
with permission.[80] Copyright 2005, Institute of Physics.

Material T range (K) ΔSmax (2T) (J kg−1 K−1) ΔT (2T) (K) TC (K) Cost ∋/kg Density (x103 kg m−3) Ref.
Gd 270–310 5 5.8[d] 293 20 7.9 [79]
[d]
Gd5Ge2Si2 150–290 27 6.6 272 60 7.5 [134]
La(Fe,Si)H 180–350 19 7[c] 300 8 7.1 [87,88]
MnAs 220–320 32 4.1[d] 287 10 6.8 [101]
MnNiGa 310–350 15 2[c] 317 10 8.2 [116]
MnFe(P,As) 150–450 32 6[d] 292 7 7.4 [134]

Adv. Mater. 2011, 23, 821–842 © 2011 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim wileyonlinelibrary.com 839
 www.advmat.de
REVIEW
40
35 ΔB: 0 - 2 T
30
25
-ΔSm(J/K-kg)
20
15
10
5 research effort in magnetic materials and computer
0 MnAs0.9Sb0.1
250 275 300 325
T(K)
Figure 25. Magnetic-entropy change of different materials at magnetic field changes of 2 T.
Reproduced with permission.[80] Copyright 2005, Institute of Physics.
applications and improved and innovated magnetic materials
will play an even more important role in energy solutions. They
contribute to saving electric power and reducing CO2 emissions
in multiple ways. For example, the conversion of electrical
energy into mechanical work and vice versa is done using elec-
tric motors and generators, respectively, which imply the use of
hard and soft magnetic materials. Permanent magnets play an
essential role in improving the efficiency of electricity transmis-
sion and utilization and also in the progressive replacement of
oil-based fuels in transportation by electric motors. For electric
vehicles, magnetic materials must retain their properties up to
moderately high temperatures, which is demanding for most
of the materials currently in use. Thus, optimizing soft and
hard magnetic materials and extending the temperature span
in which they are applicable implies a notable enhancement in
the energy efficiency of these devices.
There are also novel magnetic materials that can help us
reach larger energy efficiency. One example is magnetic refrig-
erants for solid-state energy-efficient cooling. Taking into
account that the largest electricity consumption in the domestic
market is related to refrigeration and air conditioning, the
improvement of these devices via the use of magnetic technolo-
gies will translate into a large reduction in energy consumption.
Other examples are power-MEMS for energy harvesting and
electromagnetic conversion using microfabricated permanent
magnets. Here, an electromagnetic vibration-to-electrical power
generator can efficiently scavenge energy from low-frequency
external vibrations. The generator up-converts low-frequency
environmental vibrations to a much higher frequency through
an electromechanical frequency up-converter using a magnet
and, hence, provides efficient energy conversion even at low
frequencies.
While recent progress in magnetic materials developments
have resulted in advances in increasing energy density, reducing
rare-earth contents, and lowering eddy current losses, many of
these developments require sustained research investment to
improve the magnetization, temperature stability, and operation
frequency. Investment in the development of high-efficiency
hard and soft magnetic materials now will enable more flexible
engineering designs for electronics and transportation to meet
tomorrow’s energy needs.
840 wileyonlinelibrary.com © 2011 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
www.MaterialsViews.com
Acknowledgements
Gd
Gd5Ge2Si2 O.G. thanks the DFG, BMBF, EU, and various
Gd5Ge2Si2 optimal industrial partners for their financial support in the
past twelve years to his research effort in advanced
Mn1.1Fe0.9P0.5As0.5
functional magnetic materials. M.A.W. thanks
MnFeP0.5As0.3Si0.2 NRL, ONR, and DARPA for financial support to his
MnFeP0.45As0.55 research effort in magnetic materials. E.B. thanks
the STW, FOM, and NWO for financial support.
LaFe11.4Si1.6H CHC thanks the US Air Force, ARO, DARPA, NASA,
LaFe11.4Si1.6H1.5 and NSF for their support in the past decade to her
MnAs
simulations. S.G.S. acknowledges several grants
and contracts from the US DOE, DARPA, ARO,
MnFeP0.8Ge0.2 Naval Air Systems Command, Lakehurst, NJ, and
Pennsylvania’s DEP. J.P.L. thanks the US/DoD/
DARPA, ARO, ONR, NSF, and the University of
Texas for their financial support in the past ten years
to his research efforts in nanocomposite magnetic
materials.
Received: June 14, 2010
Published online: December 15, 2010
[1] For a thorough discussion on this topic, see “U. S. Adoption of
High-Efficiency Motors and Drives: Lessons Learned: M. Lowe,
R. Golini, and G. Gereffi, published by Center on Globalization,
Governance and Competitiveness (February 2010).
[2] International Energy Annual, U. S. Energy Information Administration,
2006.
[3] Annual Energy Review, Energy Information Administration, U. S.
Department of Energy, 2007, p. 62.
[4] Annual Energy Review, Tables 2.1(b,f), 2008, pp. 41 and 45.
[5] K. A. Gschneidner, Jr., V. K. Pecharsky, A. O. Tsokol, Rep. Prog. Phys.
2005, 68, 1479.
[6] C.-J. Winter, Int. J. Hydrogen Energy 2009, 34, S1.
[7] M. Z. Jacobson, M. A. Delucchi, Sci. Am. 2009, 301, 58.
[8] O. Gutfleisch, J. Phys. D: Appl. Phys. 2000, 33, R157.
[9] S. Sugimoto, in Proc. 20th Int. Workshop on Rare Earth Permanent
Magnets and their Applications, (Ed: D. Niarchos), 2008, 106.
[10] D. Hinz, J. Lyubina, G. Fuchs, D. N. Brown, B. M. Ma,
O. Gutfleisch, K.-H. Müller, J. Magn. Magn. Mater. 2004, 272 – 276,
e321.
[11] C. H. Chen, M. H. Walmer, S. Liu, IEEE Trans. Magn. 2004, 40,
2937.
[12] O. Gutfleisch, K.-H. Müller, K. Khlopkov, M. Wolf, A. Yan,
R. Schäfer, T. Gemming, L. Schultz, Acta Mater. 2006, 54, 997.
[13] A. S. Holmes, G. Hong, K. R. Pullen, J. Microelectromech. Sys. 2005,
14, 54.
[14] H. Raisigel, O. Cugat, J. Delamare, Sens. Actuators, A 2006,
103–131, 438.
[15] S. Das, D. P. Arnold, I. Zana, J. W. Park, J. H. Lang, IEEE MEMS
2005 Conference, Miami, USA, 2005, 287.
[16] O. Gutfleisch, N. Dempsey, in: Magnetic Nanostructures in Modern
Technology, (Eds: B. Azzerboni, G. Asti, L. Pareti, M. Ghidini),
Springer, Berlin, 2008, 167.
[17] N. M. Dempsey, A. Walther, F. May, D. Givord, K. Khlopkov,
O. Gutfleisch, Appl. Phys. Lett. 2007, 90, 092509.
[18] C. H. Chen, M. S. Walmer, M. H. Walmer, S. Liu, G. E. Kuhl,
G. K. Simon, in Materials Research Society Symposium Proceedings
577, Advanced Hard and Soft Magnetic Materials, MRS, Warren-
dale PA 1999, 277.
[19] G. C. Hadjipanayis, W. Tang, Y. Zhang, S. T. Chui, J. F. Liu, C. Chen,
H. Kronmüller, IEEE Trans. Magn. 2000, 36, 3382.
Adv. Mater. 2011, 23, 821–842

www.MaterialsViews.com
[20] M. S. Walmer, C. H. Chen, M. H. Walmer, IEEE Trans. Magn. 2000,
36, 3376.
[21] C. H. Chen, Engineering magnetic materials and their applications,
Course MAT-512, University of Dayton, 2006–2010.
[22] Yang Luo, Proc. 20th Int. Workshop on Rare Earth Permanent Mag-
nets and Their Applications, (Ed: D. Niarchos), 2008, 27.
[23] US Magnetic Materials Association, http://www.usmagnetmaterials.
com/documents/usmma-presentation-general-5-08.ppt, (accessed
September 2010).
[24] J. F. Herbst, J. J. Croat, F. E. Pinkteron, W. B. Yelon, Phys. Rev. B
1984, 29, 4176.
[25] D. Kennedy, presented at Materials for Sustainable Energy,
Birmingham, UK, September 2009.
[26] Metal Pages Homepage, www.metal-pages.com (accessed
September 2010).
[27] O. Gutfleisch, I. R. Harris, J. Phys. D: Appl. Phys. 1996, 29, 2255.
[28] O. Gutfleisch, K. Khlopkov, A. Teresiak, K. H. Müller, G. Drazic,
C. Mishima, Y. Honkura, IEEE Trans. Magn. 2003, 39, 2926.
[29] M. Zakotnik, I. R. Harris, A. J. Williams, J. Alloys Compd. 2008, 450,
525.
[30] I. R. Harris, J. P. McGuiness, D. G. R. Jones, J. S. Abell, Phys. Scr.
1987, 19, 435.
[31] J. F. Gieras, M. Wing, Permanent Magnet Motor Technology Design
and Applications, Marcel Dekker, N. Y., USA 2002.
[32] Handbook of Small Electric Motors, (Eds: W. H. Yeadon,
A. W. Yeadon), McGraw-Hill, New York 2001.
[33] Jun Kang, General Purpose Permanent Magnet Motor Drive without
Speed and Position Sensor, Yaskawa Electric Corp., Japan, Yaskawa
Application Note WP.AFD.05.pdf, http://www.yaskawa.com/site/
dmdrive.nsf/link2/NKOE-7U9HJG/$file/WP.AFD.05.pdf (accessed
September 2010).
[34] Y. Honkura, Proc. of 19th Int. Workshop on Rare Earth Permanent
Magnets and Their Applications, J. Iron Steel Res. Int. 2006, 13, 231.
[35] S. Simizu, F. Pourarian, W. E. Wallace, Proc. 11th Int. Workshop on
Rare Earth Magnets and Their Applications, (Ed: S. G. Sankar), Carn-
egie Mellon University, Pittsburgh PA 1990, 191.
[36] World Wind Energy Association, http://www.wwindea.org, report
from March 2010.
[37] H. Polinder, F. F. A. Van Der Pijl, G. J. de Vilder, P. Tavner, IEEE
Trans. Energy Conversion 2006, 21, 725.
[38] S. Makaremi, Clipper’s design approach to improving reliability,
http://www.sandia.gov/wind/2007ReliabilityWorkshopPDFs/Mon-
11-ShawMakaremi.pdf (accessed September 2010).
[39] G. Bywaters, V. John, J. Lynch, P. Mattila, G. Norton, J. Stowell,
M. Salata, O. Labath, A. Chertok, D. Hablanian, Northern Power
Systems Wind PACT Drive Train Alternative Design Study Report
(2005), National Renewable Energy Laboratory, http://www.nrel.
gov/docs/fy05osti/35524.pdf, (accessed September 2010).
[40] E. de Vries, Renewable Energy World Int. Magazine 13, 2010.
[41] Clipper Windpower, http://www.clipperwind.com/pdf/liberty_
brochure.pdf (accessed September 2010).
[42] Henri Pieper, Regelungsvorrichtung für mit Dynamomaschinen
gekuppelte Explosionskraftmaschinen, Patentschrift Nr. 21202,
Kais. Königl. Österreichische Patentamt, 1905 (Pieper first sub-
mitted the German version in 1905 and four years later the US
patent).
[43] Toyota Motor Corporation, Public Affairs Division, “TOYOTA
HYBRID SYSTEM”, May 2003.
[44] C. C. Chan, Proc. IEEE 2002, 90, 247.
[45] C. H. Chen, personal communication (2009).
[46] C. Cossar, M. Popescu, T. J. E. Miller, M. McGilp, M. Olaru, IEEE
Trans. Ind. Appl. 2008, 44, 1210.
[47] K. Hirota, T. Ohashi, N. Nakamura, T. Minowa, M. Honshima,
Proc. of 20th Int. Workshop on Rare Earth Permanent Magnets and
their Applications, (Ed: D. Niarchos), 2008, 122.
Adv. Mater. 2011, 23, 821–842 © 2011 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
www.advmat.de
REVIEW
[48] W. F. Brown, Rev. Mod. Phys. 1945, 17, 15.
[49] W. F. Li, T. Ohkubo, K. Hono, Acta Mater. 2009, 57, 1337.
[50] J. P. Liu, in Nanoscale magnetic materials and applications, (Ed:
J. P. Liu, O. Gutfleisch, E. Fullerton, D. Sellmyer), Springer,
New York 2009, p. 309.
[51] E. F. Kneller, R. Hawig, IEEE Trans. Magn. 1991, 27, 3588.
[52] V. M. Chakka, B. Altuncevahir, Z. Q. Jin, Y. Li, J. P. Liu, J. Appl. Phys.
2006, 99, 08E912.
[53] Y. P. Wang, Y. Li, C. B. Rong, J. P. Liu, Nanotechnology 2007, 18,
465701.
[54] C. B. Rong, Y. Zhang, N. Poudyal, X. Y. Xiong, M. J. Kramer,
J. P. Liu, Appl. Phys. Lett. 2010, 96, 102513.
[55] Grid 2030 – A National Vision for Electricity’s Second 100 Years,
Office of Electric Transmission and Distribution, U. S. Department
of Energy, July 2003.
[56] The Smart Grid: An Introduction, Litos Strategic Communica-
tion under DOE contract DE-AC26 – 04NT41817, www.oe.energy.
gov/DocumentsandMedia/DOE_SG_Book_Single_Pages(1).pdf
(accessed September 2010).
[57] Z. Chen, F. Blaabjerg, Renewable and Sustainable Energy Reviews
2009, 13, 1288.
[58] G. M. J. Herbert, S. Iniyan, E. Sreevalsan, S. Rajapandian, Renew-
able Sustainable Energy Rev. 2007, 11, 1117.
[59] Annual Energy Review, Energy Information Administration, U. S.
Department of Energy, 2007, p. 62.
[60] C.-W. Chen, Magnetism and Metallurgy of Soft Magnetic Materials,
Dover Publications, New York 1986.
[61] R. C. O’Handley, Modern Magnetic Materials: Principles and Appli-
cations, John Wiley & Sons, New York 2000.
[62] Metglas Inc., Alloy Specification Sheets.
[63] W. Shen, PhD Thesis, Virginia Polytechnic Institute and State
University, 2006.
[64] T. McLyman, Magnetic Core Selection for Transformers and Inductors,
2nd Ed., Marcel Dekker, Inc., New York 1997.
[65] M. E. McHenry, M. A. Willard, D. E. Laughlin, Prog. Mater. Sci.
1999, 44, 291.
[66] R. Hasegawa, D. Azuma, J. Magn. Magn. Mater. 2008, 320, 2451.
[67] R. Hasegawa, J. Magn. Magn. Mater. 2000, 215–216, 240.
[68] I. Hadjipaschalis, A. Poullikkas, V. Efthimiou, Renewable Sustain-
able Energy Rev. 2009, 13, 1513.
[69] R. Hasegawa, J. Non-Cryst. Solids. 2001, 287, 405.
[70] K. Billings, Power Electron. Technol., June 2004, p. 56.
[71] W. G. Hurley, W. H. Wolfle, J. G. Breslin, IEEE Trans. Power Elect.
1998, 13, 651.
[72] K. E. Knipling, M. Daniil, M. A. Willard, Appl. Phys. Lett. 2009, 95,
222516.
[73] M. A. Willard, M. Daniil, in Nanoscale Magnetic Materials and
Applications, (Ed: J. P. Liu, O. Gutfleisch, E. Fullerton, D. Sellmyer),
Springer, New York., 2009, 373.
[74] C. Zimm, A. Jastrab, A. Sternberg, V. Pecharsky, K. Geschneidner Jr,
Adv. Cryog. Eng. 1998, 43, 1759.
[75] A. M. Tishin, Y. I. Spichkin, The Magnetocaloric Effect and Its Appli-
cations, Institute of Physics, Bristol, 2003.
[76] E. Brück, in Handbook of Magnetic Materials, vol. 17, (Ed:
K. H. J. Buschow), Publ. North-Holland, Amsterdam, 2008,
p. 235.
[77] L. Caron, Z. Q. Ou, T. T. Nguyen, D. T. C Thanh, O. Tegus, E. Bruck,
J. Magn. Magn. Mater. 2009, 321, 3559.
[78] K. A. Gschneidner, V. K. Pecharsky, A. O. Pecharsky, C. B. Zimm,
Rare Earths ‘98, 1999, 315–3, 69.
[79] O. Tegus, E. Brück, K. H. J. Buschow, F. R. de Boer, Nature 2002,
415, 150.
[80] E. Brück, J. Phys. D: Appl. Phys. 2005, 38, R381.
[81] P. I. Kripyakevich, O. S. Zarechnyuk, E. I. Gladyshevsky,
O. I. Bodak, Z. Anorg. Chem. 1968, 358, 90.
wileyonlinelibrary.com 841
 www.advmat.de
REVIEW
[82] M. Palstra, J. A. Mydosh, G. J. Nieuwenhuys, A. M. Van Der Kraan,
K. H. J. Buschow, J. Magn. Magn. Mater. 1983, 36, 290.
[83] F. X. Hu, B. G. Shen, J. R. Sun, X. X. Zhang, Chin. Phys. 2000, 9,
550.
[84] S. Fujieda, A. Fujita, K. Fukamichi, Appl. Phys. Lett. 2002, 81,
1276.
[85] O. Gutfleisch, A. Yan, K.-H. Müller, J. Appl. Phys. 2005, 97, 10M305.
[86] X. B. Liu, X. D. Liu, Z. Altounian, G. H. Tu, J. Alloys Compd. 2005,
397, 120.
[87] J. Lyubina, O. Gutfleisch, M. D. Kuz’min, M. Richter, J. Magn.
Magn. Mater. 2008, 320, 2252.
[88] A. Fujita, S. Fujieda, Y. Hasegawa, K. Fukamichi, Phys. Rev. B 2001,
65, 014410.
[89] J. Lyubina, K. Nenkov, L. Schultz, O. Gutfleisch, Phys. Rev Lett.
2008, 101, 177203.
[90] F. X. Hu, B. G. Shen, J. R. Sun, Z. H. Cheng, Phys. Rev. B 2001, 64,
012409.
[91] J. R. Proveti, E. C. Passamani, C. Larica, A. M. Gomes,
A. Y. Takeuchi, A. Massioli, J. Phys. D: Appl. Phys. 2005, 38,
1531.
[92] A. Fujita, S. Fujieda, Y. Hasegawa, K. Fukamichi, Phys. Rev. B 2003,
67, 104416.
[93] K. Mandal, O. Gutfleisch, A. Yan, A. Handstein, K.-H. Müller,
J. Magn. Magn. Mater. 2005, 290, 673.
[94] F. W. Wang, G. J. Wang, F. X. Hu, A. Kurbakov, B. G. Shen,
Z. H. Cheng, J. Phys.: Condens. Matter 2003, 15, 5269.
[95] F. X. Hu, J. Gao, X. L. Qian, M. Ilyn, A. M. Tishin, J. R. Sun,
B. G. Shen, J. Appl. Phys. 2005, 97, 10M303.
[96] J. Lyubina, R. Schäfer, N. Martin, L. Schultz, O. Gutfleisch, Adv.
Mater. 2010, 22, 3735.
[97] L. Pytlik, A. Zieba, J. Magn. Magn. Mat. 1985, 51, 199.
[98] H. Fjellvag, A. Kjekshus, Acta Chem. Scand., Ser. A 1984, 38, 719.
[99] H. Wada, Y. Tanabe, Appl. Phys. Lett. 2001, 79, 3302.
[100] H. Wada, T. Asano, J. Magn. Magn. Mater. 2005, 290, 703.
[101] H. Wada, C. Funaba, T. Asano, M. Ilyn, A. M. Tishin, Sci. Tech. Froid
C. R., 2005 2005, 2005–4, 37.
[102] N. K. Sun, W. B. Cui, D. Li, D. Y. Geng, F. Yang, Z. D. Zhang, Appl.
Phys. Lett. 2008, 92, 072504.
[103] O. Hezcko, N. Scheerbaum, O. Gutfleisch, in Nanoscale mag-
netic materials and applications, (Eds: J. P. Liu, O. Gutfleisch,
E. Fullerton, D. Sellmyer), Springer, New York 2009, p. 399.
[104] P. J. Webster, K. R. A. Ziebeck, S. L. Town, M. S. Peak, Philos. Mag.
B 1984, 49, 295.
[105] F. X. Hu, B. G. Shen, J. R. Sun, G. H. Wu, Phys. Rev. B 2001, 64,
132412.
[106] J. Marcos, A. Planes, L. Manosa, F. Casanova, X. Batlle, A. Labarta,
B. Martinez, Phys. Rev. B 2002, 66, 224413.
[107] Y. K. Kuo, K. M. Sivakumar, H. C. Chen, J. H. Su, C. S. Lue, Phys.
Rev. B 2005, 72, 054116.
[108] W. B. Cui, W. Liu, X. H. Liu, S. Guo, Z. Han, X. G. Zhao,
Z. D. Zhang, J. Alloys Compd. 2009, 479, 189.
[109] Y. Long, Z. Y. Zhang, D. Wen, G. H. Wu, R. C. Ye, Y. Q. Chang,
F. R. Wan, J. Appl. Phys. 2005, 98, 033515.
842 wileyonlinelibrary.com © 2011 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
www.MaterialsViews.com
[110] T. Krenke, E. Duman, M. Acet, E. F. Wassermann, X. Moya,
L. Manosa, A. Planes, Nat. Mater. 2005, 4, 450.
[111] T. Krenke, M. Acet, E. F. Wassermann, X. Moya, L. Manosa,
A. Planes, Phys. Rev. B 2005, 72, 014412.
[112] A. K. Pathak, I. Dubenko, J. C. Mabon, S. Stadler, N. Ali, J. Phys. D:
Appl. Phys. 2009, 42, 045004.
[113] P. J. Shamberger, F. S. Ohuchi, Phys. Rev. B 2009, 79, 144407.
[114] L. Manosa, X. Moya, A. Planes, O. Gutfleisch, J. Lyubina, M. Barrio,
J.-L. Tamarit, S. Aksoy, T. Krenke, M. Acet, Appl. Phys. Lett. 2008,
92, 012515.
[115] N. Scheerbaum, D. Hinz, O. Gutfleisch, K.-H. Müller, L. Schultz,
Acta Mater. 2007, 55, 2707.
[116] J. Liu, N. Scheerbaum, S. Weiß, O. Gutfleisch, Appl. Phys. Lett.
2009, 95, 152503.
[117] H. Fujii, T. Hokabe, T. Komigaichi, T. Okamoto, J. Phys. Soc. Jpn.
1977, 43, 41.
[118] P. Jernberg, A. Yousif, L. Hägström, J. Solid State Chem. 1984, 53,
313.
[119] O. Beckmann, L. Lundgren, in Handbook of Magnetic Materials;
Vol. 6, (Ed. K. H. J. Buschow), Publ. North Holland, Amsterdam,
1991, 181.
[120] O. Tegus, E. Brück, K. H. J. Buschow, F. R. de Boer, Nature 2002,
415, 150.
[121] E. Brück, O. Tegus, X. W. Li, F. R. de Boer, K. H. J. Buschow, Physica
B 2003, 327, 431.
[122] O. Tegus, E. Brück, L. Zhang, Dagula, K. H. J. Buschow,
F. R. de Boer, Physica B 2002, 319, 174.
[123] V. K. Pecharsky, K. A. Gschneider, A. O. Pecharsky, A. M. Tishin,
Phys. Rev. B 2001, 6414, 144406.
[124] W. Dagula, O. Tegus, X. W. Li, L. Song, E. Brück, D. T. Cam Thanh
F. R. de Boer, K. H. J. BuschowJ. Appl. Phys. 2006, 99,
08Q105.
[125] D. T. C. Thanh, E. Brück, O. Tegus, J. C. P. Klaasse,
T. J. Gortenmulder, K. H. J. Buschow, J. Appl. Phys. 2006, 99, 08Q107.
[126] D. T. C. Thanh, E. Brück, N. T. Trung, J. C. P. Klaasse,
K. H. J. Buschow, Z. Q. Ou, O. Tegus, L. Caron, J. Appl. Phys. 2008,
103, 07B318.
[127] A. Yan, K. H. Müller, L. Schultz, O. Gutfleisch, J. Appl. Phys. 2006,
99, 08K903.
[128] N. T. Trung, Z. Q. Ou, T. J. Gortenmulder, O. Tegus,
K. H. J. Buschow, E. Brück, Appl. Phys. Lett. 2009, 94, 102513.
[129] V. K. Pecharsky, K. A. Gschneidner, J. Appl. Phys. 2001, 90,
4614.
[130] M. A. Richard, A. M. Rowe, R. Chahine, J. Appl. Phys. 2004, 95,
2146.
[131] K. A. Gschneidner, V. K. Pecharsky, Int. J. Refrig. 2008, 31, 945.
[132] N. T. Trung, J. C. P. Klaasse, O. Tegus, D. T. Cam Thanh,
K. H. J. Buschow, E. Brück, J. Phys. D: Appl. Phys. 2010, 43,
015002.
[133] K. A. Gschneidner, V. K. Pecharsky, E. Brück, H. G. M. Duijn,
E. M. Levin, Phys. Rev. Lett. 2000, 85, 4190.
[134] E. Brück, M. Ilyn, A. M. Tishin, O. Tegus, J. Magn. Magn. Mater.
2005, 290, 8.
Adv. Mater. 2011, 23, 821–842